Building bridges safely and efficiently
As phobias go, the fear of bridges — gephyrophobia — is fairly common, and irrational indeed, given that actual bridge collapses like the 2007 mishap in Minneapolis, Minn., are extremely rare. The government would like to keep it that way, and beginning October 1, 2007, the Federal Highway Administration required all states to adopt Load and Resistance Factor Design (LRFD) specifications for new bridges, “to produce a uniform level of safety.”
“LRFD gives you the tools to design for a specific reliability,” said Erik Loehr, associate professor in civil and environmental engineering at the University of Missouri. “National LRFD design specifications were developed by the American Association of State Highway and Transportation Officials (AASHTO ); however, individual states are given the opportunity to develop their own specifications to address local issues and conditions as ground conditions vary substantially from state to state. AASHTO specifications adopt a conservative approach because they must take into account all possible conditions in every state.”
“In contrast, state-specific specifications need only address conditions within their borders,” Loehr explained. “State-specific guidelines can be used to gain efficiencies in design without compromising the level of safety.”
Loehr is the lead researcher for the geotechnical arm of a two-part Missouri Department of Transportation (MoDOT) comprehensive research project to develop cost-effective, technically sound bridge design specifications. Eight researchers — four from MU and four from Missouri University of Science and Technology (MS&T) — are cooperating on characterizing the effects of ground conditions in Missouri’s bridge design and construction.
“This is the largest research project we have undertaken,” said Jennifer Harper, an organizational performance engineer with MoDOT who is working with the research team. “We started discussing this in 2007, and those talks led to two basic research thrust areas: a geotechnical thrust and a structures thrust. Dr. Loehr has put in a lot of background work on it.”
The National University Transportation Center (NUTC), located in Rolla, is providing matching funds for the project.
Historically, bridges in this country have been built with an eye to generalized resistance forces rather than researched reliability factors. This built-in “cushion” relies on the judgment of individual engineers and often adds unnecessary costs to construction, something that MoDOT would like to avoid, given recent funding challenges.
“MoDOT Director Pete Rahn and Kevin Keith, MoDOT’s chief engineer, are looking for the best use of the department’s finite resources and finding opportunities to build the products they need to provide, without spending excessively,” Loehr explained.
The geotechnical work has been broken down into four tasks, and team members to date have concentrated efforts on the first task, which is site characterization. Work began in November 2008, and the sampling for this portion of the research is nearly completed.
Site characterization involves boring for soil samples — lots of them — and measuring their properties to build profiles of conditions that must be taken into account in bridge construction.
“We want to look at site characterization methods and how they affect the reliability of our bridge structures,” said Loehr. “How you drill, how you take care of samples and how you test them introduces bias and variability into design. We’re trying to quantify these effects so that MoDOT can design and construct bridges and highways more cost effectively. ”
In addition to conducting tests on soil samples in the field, teams of researchers are testing samples at both MU and MS&T to corroborate results.
“By looking at different conditions and taking the best quality measurements we can, we can improve procedures, measure how much that impacts a structure, and build it accordingly,” Loehr said.
Bill Likos, an associate professor in civil and environmental engineering, is heading up testing at MU and said that the soil sampling the team is doing has been going well. “Our objectives are right on target. The kcICON construction site (Paseo Bridge in North Kansas City) is our last site.”
Likos explained that this final site is one that is representative of areas that are primarily shale, and that other tests have been conducted at sites with different conditions that are representative of soils across the state.
“There’s no one-size-fits-all,” he added. “It is very site-specific testing.”
Back in the labs, the soil specimens are subjected to pressure until they fail.
“This is the largest lab study ever done at MS&T,” said Rick Stephenson, professor of civil engineering, one of the team members working on soil characterization, “and it’s the most cooperative project we’ve participated in. We’ve taken over 300 samples so far — in different soils and different conditions. Strength tests will simulate the load on a foundation, consolidation tests allow us to evaluate the settlement of soil and additional classification tests allow us to relate performance to other soils across the state. All soils have their own peculiarities.”
“MoDOT has been extremely accommodating in providing us with sites to investigate, drilling and sampling, and lab and field testing,” said Norbert Maerz, program head in geological engineering at MS&T. “We have a lot of data, and we’re looking at it to determine what it all means.”
Maerz explains that he also is interested in using the cone penetrometer testing method to replace traditional drilling methods for some additional project data needs. “It is an instrument that is just pushed into the soil and is more cost effective than traditional drilling,” he said.
The research team is poised to begin task two: bridge foundation load testing.
“We’re looking at various foundation types for bridges — driven piles, drilled shafts, and spread footings,” said Loehr. “The field load tests will allow us to actually measure the reliability of the foundations and then work to improve on it. MoDOT has done some testing for specific bridges in the past. We’re looking at that data and planning to perform as many as 20 additional load tests at a few different sites around the state.”
“It’s unique to be able to do this many tests,” said Brent Rosenblad, an assistant professor in civil and environmental engineering at MU. “We will begin load testing toward the first of the year. Right now we are selecting sites, purchasing equipment, and designing the test foundations, which contractors will install. We will then do the instrumentation and conduct the load tests.”
Two different types of tests are planned, Rosenblad explained. One involves drilling three shafts — a central test shaft with gauges at several points and reaction shafts on either side. The other involves jacking a pile from the bottom. In both tests, pressure is applied to the foundation until it fails, which allows for direct measurement of settlement and capacity.
“It’s forward thinking of MoDOT to fund something like this,” said Rosenblad. “The more you know, the more efficiently and confidently you can proceed.”
Also working on load testing is Ronaldo Luna, associate professor of civil engineering at MS&T whose personal research includes working with micropiles, a new bridge foundation technology. “It is smaller, more like the roots on a tree instead of one large pile,” he explained.
“Over 150 driven pile load tests have been performed all over the state,” said Luna. “More tests will help us determine the ultimate capacity to optimize foundations. If we don’t know how things will behave, design must be conservative. But with testing and data, more elegant design is possible that will not compromise safety.”
These tests will be completed within the year, at which point researchers will begin task three: establishing acceptable risk.
Using MoDOT’s data about the costs of bridge construction, researchers within the group, led by John Bowders, William A. Davidson professor in civil and environmental engineering at MU, will use the data they’ve collected on embankment and bridge design limits to begin to analyze economic risk factors. They will then make recommendations for various risk levels and the anticipated costs associated with designing for such risks.
After a review process and feedback from MoDOT, target risk levels will be established and the entire research program will culminate with task four: producing new LRFD design specifications for Missouri. Results of this large cooperative project are scheduled for completion a year from now on Nov. 1, 2010.
Implementation of the new specifications are intended to produce more effective designs and substantial cost savings, not just in bridge design, but also in future decisions about site investigation and design testing. It is expected that Missouri’s new specifications will become a national model as other states look to personalize LFRD specifications.
“With the results from this research program MoDOT can make rational design decisions based on the most up-to-date information available,” said Loehr. “This research is about getting the reliability we need from our transportation infrastructure while at the same time making the most effective use of the taxpayer dollar possible.”